Design and Testing of an MPPT Algorithm Using an Intelligent RBF Neural Network and Optimum Relation Based Strategy

Extracting maximum possible power from wind energy system and deliver of energy to the grid proficiently requires the generator converters to be properly tuned. This article employs an adaptive network-based radial basis function (RBF) artificial neural network (ANN) algorithm to track the optimum rotor conditions of the generator under the changing wind velocity conditions. The single ANN based algorithm rapidly adjusts the generator-side converter reference speed settings. An optimum relation based MPPT algorithm determines the grid-side converter dc voltage reference settings. The algorithms do not require any anemometer for wind speed measurement or the system parameters determination. The difference in actual values and set values of speed and dc voltage, respectively, drive the PI controllers to regulate the modulation indices of the converters for maximum system efficiency. The MPPT algorithms were verified on a permanent magnet synchronous generator characterized through a detailed dynamic model of the generator, the drive train and the converter systems. Simulations results confirm that the wind energy system able to supply power to the transmission line keeping maximum power coefficient for arbitrarily changing the operating situations.